Your genome is not a static instruction manual. It is a choreography.
For decades, the standard model treated the three-dimensional structure of DNA as something that was largely fixed, a scaffold that determined which genes were accessible and which were not. A paper published in Nature Genetics on February 16 overturns that assumption. Researchers at the Salk Institute for Biological Studies found that the genome is in constant motion, continuously folding and unfolding at active genes — and the speed of that motion is itself functionally important. Genes that are turned on tend to be surrounded by loops that form and dissolve quickly. Genes that are silenced sit inside loops that hold their shape for hours.
The finding, led by Tessa Popay, a postdoctoral researcher, and Jesse Dixon, who holds the Helen McLoraine Developmental Chair at Salk, has implications for both cancer and autism. When the machinery that controls genome folding goes wrong, cells can lose track of who they are supposed to be.
"The continuous folding and unfolding of our genome may be particularly important for helping a cell remember who it is supposed to be," Dixon said in a Salk press release.
Loop formation is mediated by a protein complex called cohesin, which rides along DNA like a ring sliding down a cord, pinching the molecule into loops. Cohesin's movement requires an accessory protein called NIPBL. When the researchers depleted NIPBL in human cells, cohesin could no longer move efficiently. The genome did not fold uniformly. Some regions lost their loops within minutes. Others held on for many hours before unwinding.
The asymmetry was consequential. In cells depleted of NIPBL during the transition out of cell division, 549 genes changed their expression — 457 of them decreased. The researchers classified 16,860 chromatin loops as cohesin-dependent based on a complementary depletion of a cohesin subunit called RAD21. The loops that turned over most rapidly were associated with active gene expression. The stable loops, the ones that took hours to unravel, were associated with silent genome regions.
The pattern was tissue-specific. In heart cells, the dynamic folding was most pronounced at genes related to heart function. In neurons, it was most pronounced at genes related to neuronal function. This aligns with what happens when mutations disrupt the folding machinery: the result is not a single organ problem but a syndrome affecting different body parts in different ways. Cornelia de Lange syndrome, which involves malformations of the limbs, face, and digestive system, is linked to mutations in the cohesin-NIPBL system.
The cancer connection runs through the same principle of cellular identity. A healthy cell maintains the expression of genes that define its type. Cancer disrupts that maintenance, allowing cells to ignore their molecular identity and proliferate unchecked.
"Cancer is potentially exploiting that same principle, changing where in the genome these dynamics are more important to manipulate cell identity and encourage uncontrolled growth," Dixon said.
The discovery is a return to first principles for a field that had settled prematurely into a static model. Every human cell contains roughly two meters of DNA carrying about six billion base pairs. Packing that length into a nucleus requires extensive looping, and the loop architecture was known to vary between cell types. What was less appreciated was that the same loops do not persist indefinitely, even within a single cell type. The structure is dynamic at a timescale that is biologically meaningful, not just a consequence of thermal motion.
The finding also raises the question of whether genome folding dynamics are targetable. If cancer is exploiting a process that healthy cells use to maintain their identity, drugs that modulate loop dynamics could in principle reset cellular behavior. That is a long way from a therapeutic lead, but it is the kind of mechanistic question that pharma watches closely after a clean basic science result. The paper appeared in Nature Genetics on February 16. The research was supported by federal funding and private sources at Salk.
